Tuesday, January 31, 2023
processing....
At the American Society of Clinical Pathology (ASCP) 2007 Annual Meeting, Antonia R. Sepulveda, MD, PhD,[1] from the Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, gave a lecture on microsatellite instability (MSI) testing in colonic neoplasia.
MSI testing in tumors has been steadily implemented in clinical practice for evaluation of colon cancer. However, we are still far from being able to state that this is a test with generalized use, implying that many patients who would benefit from this test will still go under the radar. It continues to be important to increase awareness of the role of MSI testing in the clinical care of patients and family relatives of patients with colon cancer.
Because pathologists routinely receive colon cancer tissues for diagnosis and cancer staging in their diagnostic practice, they can request microsatellite tests and determine the underlying DNA mismatch repair deficiency by immunohistochemistry (IHC). Therefore, pathologists should be familiar with the indications, interpretation, and limitations of MSI tests.
The main goals of Dr. Sepulveda's presentation on MSI testing of colon cancer were to:
The following is a report of highlights of the ASCP 2007 meeting, as well as additional information on MSI testing and MMR protein expression in colonic polyps, and more.
MSI is a type of mutation that occurs in small DNA segments that consist of repetitive nucleotides of generally 100-200 base pairs in length. The nucleotide sequences within the repetitive elements include mononucleotide repeats of adenine (A)n or cytosine-adenine (CA)n dinucleotide repeats. During DNA replication, these repetitive sequences may decrease or increase in length because of strand slippage. Such changes in length of the nucleotide repeat are termed microsatellite instability. In cells with deficient DNA mismatch repair, as occurs in patients with hereditary nonpolyposis colorectal cancer (HNPCC), also known as Lynch syndrome, these mutations are not repaired and persist in the mutated tumor cells. Clonal expansion of cells with this type of mutation in tumors leads to highly homogeneous cell populations, and this mutation can be easily detected in the tumor DNA by polymerase chain reaction (PCR) methods. One will scrape off an area of tumor and an area of nonneoplastic tissue from colonic tissue sections from the same patient. Several unstained paraffin tissue sections can be used to collect enough DNA for testing. The size of a standard set of microsatellite markers from tumor DNA as compared with DNA from nonneoplastic tissue from the same patient is determined by electrophoresis of DNA products amplified by PCR. The most widely used set of microsatellite markers was recommended by a National Cancer Institute (NCI) consensus group and consists of 2 mononucleotide repeat markers: BAT25 and BAT26; and 3 dinucleotide repeat markers: D2S123, D5S346, and D17S250.[2]
The identification of MSI in tumors indicates a severe deficiency of DNA mismatch repair, directly resulting from loss of one of the DNA MMR proteins. Generally there is an actual loss of a specific DNA MMR protein in tumor cells, either because of loss of gene expression or because of genetic mutations that result in unstable truncated proteins. The DNA MMR proteins consist of hMLH1, hMSH2, hMSH6, hPMS2, hMSH3, and hMLH3.[3] The most efficient and currently used approach to determine which DNA MMR protein is missing in the tumor is IHC. MSI is characteristically detected in the cancer tissue of almost all patients with HNPCC. However, it is important to be aware that approximately 15% of colorectal cancers can have MSI-positive status.[4,5]
The results of the MSI test using the NCI panel of 5 microsatellite markers are reported as MSI-High, MSI-Low, and microsatellite stable (MSS). MSI-High level (MSI-H) tumors demonstrate MSI in at least 2 of the 5 markers, MSI-Low (MSI-L) tumors show MSI in only 1 marker, and no instability is detected in any of the 5 markers in MSS tumors.
Recently, alternative panels of microsatellite markers have been used by some laboratories. In a small study of 11 MSI-H colon cancer cases, the MSI Analysis System (Promega Corp.), which includes 5 mononucleotide and 2 pentanucleotide repeat markers, showed complete concordance with the NCI-recommended panel of 5 markers for MSI-H and MSS tumors.[6] In contrast, all MSI-L cases by the NCI panel were MSS when tested by the Promega MSI Analysis System .[6] Further studies to evaluate the Promega panel performance in the detection of MSI-L cases are needed to determine its potential role in identification of the rare MSI-L cases with DNA MMR gene mutation and/or loss of function vs those without alterations in DNA MMR.
Confirmation of HNPCC requires identification of germline mutations (detected in peripheral blood DNA) in one of the DNA MMR genes.[3] In contrast, somatic hypermethylation of the hMLH1 promoter leading to loss of expression is the predominant underlying abnormality causing MSI in sporadic tumor tissues.[7]
The primary indication of MSI testing of colorectal cancer is to stepwise evaluate a patient for HNPCC. Given that a patient with MSI-H tumor will most likely be recommended to have a germline test performed, the patient and the clinician taking care of the patient need to be aware of the importance of undergoing genetic counseling. This is the case not only if the MSI test indicates the possibility of HNPCC, but even in cases in which the test is negative when there is a high suspicion of HNPCC on the basis of family history. The MSI test is performed on tumor tissue and identifies somatic mutations. The MSI test detects mutations that are acquired in tumor cell DNA during tumor progression. In contrast, germline mutations are inherited mutations affecting a given gene that is present in all cells of the body. Therefore, germline mutations can be identified in DNA extracted from blood cells. The identified germline mutation in DNA MMR genes confirms the diagnosis of HNPCC and can be used as a blood DNA test to screen family relatives to determine whether they are mutation carriers. As recommended by the NCI guidelines,[5] after a tumor is classified as MSI-H, and the specific DNA MMR gene is identified by IHC, germline mutation analysis of the DNA MMR gene that showed loss of expression in the tumor tissue is indicated. The germline mutation analyses of DNA MMR genes in the peripheral blood DNA represents a genetic test and requires patient consent. The guidelines also state that "for families with a strong suspicion of HNPCC, germline testing should be considered, even when the MSI/IHC results indicate MSI-L, MSS or normal [DNA MMR gene] expression.[5]" The DNA MMR genes most commonly affected by germline mutations in HNPCC are hMLH1 and hMSH2 (40% of the cases each), followed by hMSH6 (10% of the cases) and hPMS2 (approximately 5% of the cases).[3] As discussed below, there is a high correlation between MSI-H status and loss of expression of 1 of the 4 DNA MMR proteins. Therefore, after determining that a tumor is MSI-H, IHC is the preferred method to determine the specific protein that is lost in tumor tissue and, in turn, the gene encoding that protein can then be specifically evaluated for germline mutation. The significance of MSI-L is not well established; however, it is known that some tumors of HNPCC patients, such as cases of hMSH6 gene mutation, may show loss of that DNA MMR protein while the tumor only shows MSI-L or even MSS.[5] In most cases in which the tumor shows MSI-L, no abnormality in the main DNA MMR genes is found.[8]
In addition to HNPCC, MSI-H status is present in a proportion of sporadic colonic adenocarcinomas with unique clinical characteristics. Of note, MSI-H sporadic colorectal cancers are associated with loss of expression of hMLH1 caused by promoter hypermethylation rather than gene mutation. Samowitz and colleagues[9] reported a relationship between MSI, as determined by instability at the BAT26 marker, and improved prognosis with an approximately 60% decrease in death attributable to MSI-positive colon cancers compared with MSS colon cancers. In patients with stage III colorectal cancer, MSI (as defined by deletions in the BAT-26 mononucleotide repeat) was associated with fewer lymph node metastases and the lower nodal burden was a prognostic factor for improved survival compared with MSS tumors.[10,11] Patients with stage III MSI-H colorectal cancer were reported to have an overall increased 5-year survival (72% vs 35%) and increased relapse-free rates (66% vs 39%)[12] A growing number of studies shows that the underlying MMR deficiency in MSI-positive tumors may also have implications for tumor therapy because tumors with MSI may show increased resistance to chemotherapeutic agents, including alkylating agents, cisplatin, doxorubicin and 5-fluorouracil.[13-15] As pointed out by Parc and colleagues,[16] the MSI status of tumors may be used to determine whether chemotherapy should be given. In their study, they evaluated patients who had a curatively resected T3N0M0 colon cancer and were not receiving adjuvant therapy. Patients with MSI tumors (defined as having instability by immunohistochemical determination of loss of expression of hMLH1 or hMSH2) had better recurrence-free survival than those with MSS tumors.[16] This study suggested that prognosis for MSI-H T3N0M0 colon cancers is excellent and chemotherapy should not be proposed in these patients.
Presently, the recommended use of the MSI test is to help identify HNPCC. The revised Bethesda guidelines describe the individual criteria for selection of patients who are recommended to have MSI and/or DNA MMR expression tested in their cancer tissues,[5,17,18] as follows[5,18]:
Pathologists can significantly contribute to the selection of cases for MSI testing, particularly by requesting reflex testing in cancers of patients who meet 1 of the first 3 guidelines. MSI testing in cases that demonstrate MSI-histopathologic features in patients between the ages of 50 and 60 years identifies a high number of MSI-H cases. In a retrospective study, we found MSI-H in 57% of the cancers with MSI-histology in patients between 50 and 60 years.[19] Some laboratories are providing the MSI test results integrated with the surgical pathology report; this may become a trend in clinical practice.
Most laboratories offer immunohistochemical testing for the main DNA MMR proteins (hMLH1, hMSH2, and hMSH6), and a few laboratories also perform IHC for hPMS2, which is the underlying cause of HNPCC in a small number of cases. By performing IHC for the 4 genes, it can be easily determined which gene is most likely to be the gene with underlying genetic mutation. To interpret the IHC results, it is helpful to understand the biochemistry of DNA mismatch repair. The DNA MMR genes encode the MutS proteins hMSH2, hMSH3, and hMSH6, and the MutL proteins hMLH1, hPMS1, hPMS2, and hMLH3.[20] The stability of DNA MMR requires the formation of specific MutS and MutL protein heterodimers. The proteins undergo degradation if they cannot form stable heterodimers, which happens in HNPCC tumor tissues, because one of the critical proteins is lacking. The hMSH2 protein interacts with either hMSH6, forming the MutS-alpha complex, or with hMSH3, forming the MutS-beta complex. MutL heterodimers include MutL-alpha (hMLH1 and hPMS2), which appear to be responsible for most of the DNA MMR, and MutL-beta (hMLH1 and hPMS1), which does not appear to be significantly involved in the DNA mismatch repair.[21,22] As summarized in the Table (personal observation and Herman and colleagues[ 8]), for example, if the primary DNA MMR gene deficiency (resulting from genetic mutation) occurs in hMSH2, then IHC would reveal loss of expression of hMSH6 and hMSH3 because these genes are degraded in the absence of hMSH2. In contrast, if the primary DNA MMR gene deficiency occurs in hMSH6, then hMHS2 expression is still preserved because it can be stabilized by forming heterodimers with hMSH3.
The performance of MSI testing in combination with determination of the underlying deficient MMR gene by immunohistochemical stains of tumor sections to identify patients with HNPCC has been evaluated in several studies. Terdiman and colleagues[23] reported loss of hMSH2 or hMLH1 protein expression by IHC in 94% of tumors from patients with germline mutations or gene sequence variants. Halvarsson and colleagues[24] reported that loss of expression of at least 1 of the 3 MMR proteins tested (hMSH6, hMLH1, and hMSH2) was found in 92% of MSI tumors. In this study, of the MSI-high tumors, 8% showed normal MMR protein expression and all microsatellite-stable or MSI-low tumors showed normal immunostaining for all 3 proteins. Overall, in this study, immunostaining for the MMR proteins hMLH1, hMSH2, and hMSH6 had a sensitivity of 92% (some MSI-H tumors did not show loss of either hMLH1, hMSH2, or hMSH6 in the tumor) and a specificity of 100% (all tumors with loss of expression of either hMLH1, hMSH2, or hMSH6 were MSI-H tumors) for detecting MMR-deficient tumors. These findings support the notion that both MSI and MMR protein immunostaining should be performed to effectively identify patients likely to have HNPCC and who might have indication for germline mutation analysis.[24] In summary, most CRC with MMR protein loss show MSI-H in cancer tissues, but there are exceptions. Conversely, almost all MSI-H CRC show loss of main DNA repair proteins, but there are exceptions. A limitation of IHC includes the occasional tumor that shows unsatisfactory IHC, making the results difficult to interpret, with possible false-positive or false-negative results. Of note, generally, in typical MSI-H tumors, the loss of DNA repair protein expression in tumor cell nuclei by IHC is uniform throughout the tumor (personal observation and Hatch and colleagues[8]). Rare cases of MSI-positive tumors, detected by BAT26 instability, were reported to show heterogeneous staining patterns.[25]
| IHC MLH1 |
IHC PMS2 |
IHC MSH2 |
IHC MSH6 |
|
|---|---|---|---|---|
|
MLH1 Mutation |
Loss | Loss | Preserved | Preserved |
|
MSH2 Mutation |
Preserved | Preserved | Loss | Loss |
|
MSH6 Mutation |
Preserved | Preserved | Preserved | Loss |
|
PMS2 Mutation |
Preserved | Loss | Preserved | Preserved |
Consensus hasn't been reached in regard to whether testing a colorectal cancer with the aim of searching for HNPCC should employ a combination of MSI-test and IHC as first line tests or whether 1 of these 2 tests should be performed first, with the other being performed only in selected cases. As discussed above, the most accurate results can be obtained when the 2 tests are performed on the same tumor. This author favors the combination of MSI test and IHC evaluated and reported as an integrated pathology report. However, for some laboratories and individual patients, it may not be practical or cost-effective to perform both tests. If only IHC or the MSI test is performed and is not supportive of HNPCC, it is important to note in the report that if HNPCC is suspected, both tests should be performed; further genetic testing may be indicated if there is suspicion of HNPCC by family history, even when the IHC and MSI tests are negative.
Germline mutations of hMSH6 are associated with a minority of HNPCC cases. Somatic mutations of hMSH6 are also present in sporadic MSI-H cancers because the hMSH6 gene coding region contains target sequences for MSI-type mutations, undergoing mutagenesis during progression of MSI-H colon cancers due to primary deficiency of other MMR proteins such as hMLH1 and hMSH2.[24, 26-33] The majority of tumors in hMSH6 HNPCC mutation carriers are MSS or MSI-L, and immunohistochemical stains show the presence of hMSH6 protein in some cases with a mutation-positive gene. Hendriks and colleagues[30] reported that immunohistochemical stains in tumors had a sensitivity of 90% in predicting a mutation in hMSH6. Overall, the available data suggest that in all patients suspected to have HNPCC, hMSH6-mutation analysis should be considered if alterations in hMLH1 and hMSH2 are excluded, even if MSI testing and IHC are negative.[34]
MSI can be a molecular feature of colonic adenomas and hyperplastic polyps (MSI-H or MSI-L) in sporadic cases,[35-39] as well as in patients with HNPCC.[40,41] Generally, MSI-H is rarely detectable in polyps that do not have high-grade dysplasia.
However, some HNPCC adenomas < 0.7 cm had detectable loss of MMR protein expression even in the absence of high-grade dysplasia.[42] MSI was demonstrated in 80% of adenomas in patients with HNPCC,[40] including MSI-H in 53% and MSI-L in 27% of the cases. A significant association was found between MSI-H and high-grade dysplasia in adenomas. These data indicate that the underlying DNA MMR defect is involved in the early stages of development of HNPCC adenomas,[40,42] and it was recommended that immunohistochemical staining/MSI testing of large adenomas with high-grade dysplasia in young patients (younger than 50 years) be performed to help identify patients with suspected HNPCC.[42]
In sporadic colorectal cancer and polyps, the most frequent mechanism of MSI is silencing of the hMLH1 gene by promoter methylation and this is associated with partial or complete absence of expression of hMLH1 protein that can be demonstrated by IHC.[36] Overall, studies report a very low percentage of MSI in sporadic colonic adenomas, and, as determined in the studies by Samowitz and colleagues,[35,43] 80% of the sporadic adenomas were MSS and 18% were MSI-L. Currently, a consensus regarding which polyps should be tested for MSI and IHC for DNA MMR proteins has not been reached. However, the available data suggest that the polyps that are more likely to demonstrate an underlying MMR gene deficiency and MSI are the larger adenomas with high-grade dysplasia and the serrated adenomas.
Given current knowledge, the recommended algorithm to determine whether a patient with colorectal cancer has HNPCC/Lynch syndrome includes MSI testing and IHC for the DNA MMR proteins. Genetic mutational analysis is then performed for the gene encoding the protein shown to have loss of expression by IHC.[44]
